AMBIO (2010) 39:504514 DOI 10.

1007/s13280-010-0072-x

REPORT

Estimation of Nitrogen and Phosphorus in Efuent from the Striped Catsh Farming Sector in the Mekong Delta, Vietnam
Sena S. De Silva, Brett A. Ingram, Phuong T. Nguyen, Tam M. Bui, Geoff J. Gooley, Giovanni M. Turchini

Received: 12 September 2009 / Revised: 24 April 2010 / Accepted: 3 May 2010 / Published online: 6 July 2010

Abstract In this study an attempt is made to estimate nitrogen and phosphorus discharged to the environment from the striped catsh (Pangasianodon hypophthalmus) farming sector in the Mekong Delta (8330 10550 N, 104300 106500 E), South Vietnam. The sector accounted for 687,000 t production in 2007 and 1,094,879 t in 2008, with over 95% of the produce destined for export to over 100 countries. Commercial and farm-made feeds are used in catsh farming, currently the former being more predominant. Nitrogen discharge levels were similar for commercial feeds (median 46.0 kg/t sh) and farm-made feeds (median 46.8 kg/t sh); whilst, phosphorus discharge levels for commercial feeds (median 14.4 kg/t sh) were considerably lower than for farmmade feeds (median 18.4 kg/t sh). Based on the median nutrient discharge levels for commercial feeds, striped catsh production in the Mekong Delta discharged 31,602 t N and 9,893 t P, and 50,364 t N and 15,766 t P in 2007 and 2008, respectively. However, the amount of nutrients returned directly to the Mekong River may be substantially less than this as a signicant proportion of the water used for catsh farming as well as the sludge is diverted to other agricultural farming systems. Striped catsh farming in the Mekong Delta compared favourably with other cultured species, irrespective of the type of feed used, when the total amounts of N and P discharged in the production of a tonne of production was estimated. Keywords Efuent discharge Mekong Delta Nitrogen Nutrient mass balance Pangasius Phosphorus Striped catsh Tra Catsh BMPs

INTRODUCTION Aquaculture or farming of aquatic organisms is a very old tradition, thought to have originated in China over

2500 years back. Over the last few decades aquaculture is seen as the means of meeting the global aquatic food sh needs, consequent to the plateauing (FAO 2008) of the traditional food sh supplies from wild capture sheries. It is estimated that currently nearly 50% of the food sh consumed is from aquaculture, which is also reputed to be the fastest growing primary food production sector (Subasinghe et al. 2009). Aquaculture being a relatively new food production sector, in global terms, has been subjected to a greater degree of public scrutiny and debate (De Silva and Davy 2009). Consequently, many issues relating to its environmental impacts (Naylor et al. 1998, 2001; World Wild Life Mediterranean Programme 2005; De Silva and Turchini 2008), resource usage (Wijkstrom and New 1989; New 1991, 1997; Naylor et al. 2000, 2009) and indeed human health impacts (Sapkota et al. 2009) have been raised. It is often alleged that the ecosystem health of most tropical rivers is in a relatively poor state, primarily due to anthropogenic developments in their catchments, damming and untreated efuent discharges (Dudgeon 2000; Sodhi et al. 2004). The Mekong Delta (8330 10550 N, 104300 106500 E), South Vietnam has an area of approximately 3.9 9 106 ha with an estimated population of 17 million and, being the food basket of Vietnam, is heavily impacted by anthropogenic activities. The Mekong River, which divides into two main branches (Tien Giang and Hau Giang) as it enters the Delta, consequently receives a large quantity of efuent. Over the last decade or so the Mekong Delta has also become the home to the rather explosive farming sector of the native catsh (Pangasianodon hypophthalmus Sauvage), locally known as striped, tra and/or sutchi catsh, but with many different trade names around the world (Nguyen and Oanh 2009; Phan et al. 2009). Also this sector

123

Royal Swedish Academy of Sciences 2010 www.kva.se/en

AMBIO (2010) 39:504514

505

has recorded the highest growth rate in volume compared to any other aquaculture commodity globally over the last decade or so. The sector accounted for 687,000 and 1,094,879 t production, in 2007 and 2008, respectively (Department of Aquaculture 2008), the latter amounting to 34.3% of the total aquaculture production in Vietnam, the fth ranked nation in global aquaculture production (FAO 2008). The sector also provides employment to about 170,000 persons, mostly women (Sub-Institute for Fisheries Economics and Planning in Southern Vietnam 2009). Moreover, the great bulk of the produce is destined for export to over 100 countries (Globesh 2009), and consequently the sector has drawn the attention of many stakeholders, such as importers, environmentalists, entrepreneurs, etc. and is a commodity that standards are being developed for (Corsin 2007; Neubacher 2009). Striped catsh culture practice in the Mekong Delta, a modernised version of an old practice (Nguyen and Oanh 2009; Phan et al. 2009), stands out in many respects from
Fig. 1 The location of the main striped catsh farming areas in the Mekong Delta, Vietnam, in relation to the Delta as a whole (adopted from Phan et al. 2009). The arrow indicates the area of catsh farms that are shown in Fig. 2

other aquaculture systems and practices. Foremost, the average production ranges from 350 to 400 t/ha/crop, with sh being harvested at 11.5 kg size for processing into llets for export (Nguyen and Oanh 2009; Phan et al. 2009). Striped catsh exports worldwide have dramatically grown in recent years because it is seen as an acceptable substitute to other white sh, such as Atlantic cod (Gadus morhua L.) in Europe and channel catsh (Ictalurus punctatus) in the USA, but much less costly (Globesh 2009). Hence, striped catsh farming is one of the most important aquaculture sectors of the current global aquaculture industry. Striped catsh culture is predominantly in deep earthen ponds (average 44.5 m), mostly along the two main branches (Tien Giang and Hau Giang) of the Mekong River (Fig. 1), and are densely located (Fig. 2), often of uniform pond size, and with all farms extracting and discharging water from and into the river or connected channels. In aquaculture operations, feed is often regarded as both, the highest recurring monetary cost (Goddard

Royal Swedish Academy of Sciences 2010 www.kva.se/en

123

506 Fig. 2 A Google image showing the concentration of catsh farms. Note the uniformity of pond size. The farm area shown here is indicated by the arrow in Fig. 1

AMBIO (2010) 39:504514

1996) and the main factor responsible for potential detrimental impacts on the environment primarily through the discharge of nitrogen and phosphorus as excreta and other metabolic products (De Silva and Anderson 1996). Accordingly, this article attempts to estimate the nitrogen and phosphorus discharge into the Mekong River from the striped catsh farming sector, which occurs along the banks of the river and its channels. This farming sector was chosen in view of the fact that, in the last decade, it has recorded the highest growth rate in volume compared to any other aquaculture commodity globally, and also provides an opportunity to assess the total discharge from an aquatic farming system as a whole. In fact, and interestingly, the production from the striped catsh farming sector is concentrated in a relatively small geographical area (Figs. 1, 2), and characterised by highly homogenous farming practices; both criteria that have been rarely witnessed for any other aquaculture sector.

detailed information), proximate analysis of representative feed samples, both commercial and farm-made feeds, and information from the literature on nitrogen and phosphorus retention in the body. During this survey it was apparent that catsh farmers predominantly use commercial, pellet feeds, and a small percentage make their own feed, referred to as farm-made feeds, using locally available ingredients. Also during the survey details on feeding practices and the type of commercial feeds used at different stages of the growth cycle were obtained. This study did not compare the composition and the efcacies per se of farm-made feeds against the commercially manufactured pellet feeds, which by itself warrant a separate detailed investigation. However, it has been demonstrated that the difference in the proximate composition of the two feed types are marginal, but criteria such as water stability, digestibility, etc. between the two types are unknown (Phan et al. 2009). Feeds

MATERIALS AND METHODS In this study a number of data sources were utilised; data on sh feeding practices and the average FCR (amount of food dispensed as fed/increase in biomass of stock in wet weight) obtained by the recent survey of 97 catsh farms in the Mekong Delta implemented by Collaboration for Agriculture and Rural Development (CARD) program between the Governments of Vietnam and Australia, funded through the Australian Agency for International Development (AusAID) (see Phan et al. 2009 for more

A total of 12 commercial feed samples, from six manufacturers were chosen randomly for analysis, out of a total of about 20 relatively large feed manufacturers (each producing over 200 t/year) operating in the Mekong Delta, and catering almost exclusively to the catsh farming sector (Nguyen and Oanh 2009; Phan et al. 2009). In addition, four farm-made feeds (FMF) were obtained from different sites. Farm-made feeds use a variety of locally available ingredients, foremost among which dried sh powdered, rice bran and soy bean meal (Phan et al. 2009). On the other hand, the ingredient makes up and the proportions

123

Royal Swedish Academy of Sciences 2010 www.kva.se/en

AMBIO (2010) 39:504514

507

used in commercial feeds are not publicly available. The only information available with regard to the latter is the proximate composition which is clearly presented on the feed bags, usually each of 2025 kg (Phan et al. 2009). It is estimated that only about 3% of farmers use farm-made feeds, of which about 49% are made on site, the rest being purchased from other farm sites (Phan et al. 2009). Feed samples were collected from farms, of batches of feed that had been procured most recently and were well within the recommended shelf life period. New feed bags were opened and a sample of 500800 g, from each feed bag, for three randomly selected bags for each type of commercial feed obtained and transferred into plastic containers and secured air tight. Samples were transported to the laboratory and kept at 4C until further analysis. The specications provided in each bag were noted. The commercial feeds sampled were of three types, recommended for use for grow-out of catsh of three size groups, as specied by different feed manufacturers. The three types recommended for use for different stock sizes were 14150, 20200 and over 500 g. For convenience and clarity these feeds are termed small feed (SF), medium feed (MF) and large feed (LF), respectively, in the text and the corresponding tables and gures. Proximate composition analysis was conducted using standard procedures; moisture content by drying at 80C to constant weight, protein by Kjeldahl nitrogen method (N 9 6.25) using an automated Kjeltech (Model 435 and 323, Buchi, Switzerland), lipid by ether extraction using an automated Soxhlet extraction (Model 810, Buchi, Switzerland), and ash by combustion at 600C for 3 h in a mufe furnace (Model A-550, Vulcan, USA). Gross energy content was estimated by combusting in an oxygen atmosphere in a Bomb Calorimeter (Model 1261, Parr, USA). Phosphorus content of feeds was determined photometrically by conversion to phosphomolybdovanadate (Model Spectronic, Genesys 2, Milton Roy, USA).

Estimation of Nitrogen and Phosphorus Discharge Nutrient-balance models have been used to estimate farm nutrient discharge of N and P for several sh species, and are considered to be a more reliable and cost-effective than direct measurement on farm (Cho et al. 1991, 1994; Kelly et al. 1996; Cho and Bureau 1998; dOrbcastel et al. 2008). In the absence of detailed information on fate on nutrients in the catsh pond environment, and as we were attempting to estimate for the whole sector concentrated in the Mekong Delta, amounts on nutrients lost to sediments or removed by pond cleaning were not taken into account. In this analysis the nutrient-balance model used information on nutrients inputs (feed added) and nutrients removed (harvested sh)

to estimate gross nitrogen (N) and phosphorus (P) discharge levels from the striped catsh farming sector in the Mekong Delta. The model utilised published information on N and P content of striped catsh (Yi et al. 2004) combined with FCR data (Phan et al. 2009), and proximate analysis of feeds from the current study to estimate discharge levels. All output variables from the model are presented as kilogram of total N and total P per tonne of sh produced. Since the proximate composition of commercial feeds varied according to the size of sh being fed, the model used two production phases taking these feeding strategies of the catsh farming practices into consideration; Phase I for diets fed to sh up to 200 g and Phase II for diets fed to sh greater than 200 g. Because information on the sh biomass and the amount of feed used in each phase are not known, three different model simulations were conducted with 40, 50 and 60% of N and P loading being attributed to Phase I, and, accordingly, 60, 50 and 40% of N and P loading being attributed to Phase II. In addition, since N and P content of commercial feeds varied between manufacturers and the N and P content of farm-made feeds were highly variable, model simulations were undertaken using high (95th percentile), median and low (5th percentile) of N and P feed content values, which are detailed in Table 1. FCR data for commercial feeds and farm-made feeds were obtained from 83 and 12 farms, respectively, during a survey of catsh farms in the Mekong Delta (Phan et al. 2009). FCR values ranged from 1.0 to 3.0 (median 1.69) and 1.3 to 3.0 (median 2.25) for commercial and farmmade feeds, respectively; and, as described above, simulations were then conducted at high (95th percentile), median and low (5th percentile) FCR for each feed type. It has been suggested that in normal aquaculture practices less than 5% to as much as 30% of sh feed, in the form of dust or uneaten pellets, is not consumed by sh (Cho et al. 1991). However, this loss is taken into account in the nal FCR values and was not built into the computations. Published information on N and P content of whole striped catsh is limited. Yi et al. 2004, provided values of 6 and 0.7% (dry matter basis) for N and P, respectively, of farmed striped catsh (1.8 and 0.21% wet weight, respectively). These values were used to estimate nutrient discharge loadings for striped catsh farming.

RESULTS The commercial pellet feeds for striped catsh grow-out collected and analysed in this study are thought to be representative of the feeds available and commonly used on farms. These feed were recommended by different manufacturers as feeds suitable for sh of 14150 g (SF), 20200 g (MF) and over 500 g (LF). However, the rst

Royal Swedish Academy of Sciences 2010 www.kva.se/en

123

508

AMBIO (2010) 39:504514

Table 1 N and P nutrient loadings for different feed types, N and P content of feeds, FCR values and production phases (commercial feeds only) for striped catsh farming in the Mekong Delta (calculations used N and P content of striped catsh from Yi et al. 2004) N & P content in feed (% wet weight) Commercial feeds Low (3.60 N, 0.70 P) Median (4.34 N, 0.96 P) High (4.86 N, 1.10 P) Low (3.60 N, 0.70 P) Median (4.34 N, 0.96 P) High (4.86 N, 1.10 P) Low (3.60 N, 0.70 P) Median (4.34 N, 0.96 P) High (4.86 N, 1.10 P) Farm-made feeds Low (1.25 N, 0.51 P) Median (2.90 N, 0.92 P) High (3.39 N, 1.28 P) 60% Phase I 40% Phase II 50% Phase I 50% Phase II 40% Phase I 60% Phase II Production phase/ structure Nutrient loading (kg/t sh) at different FCR values N Low FCR (1.53) 33.4 38.2 45.4 34.0 39.9 47.2 34.5 41.5 49.0 Median FCR (1.69) 38.9 44.2 52.1 39.5 46.0 54.1 40.1 47.7 56.0 High FCR (2.0) 49.4 55.6 65.1 50.1 57.8 67.4 50.8 59.9 69.7 P Low FCR (1.53) 9.9 12.9 14.7 9.7 12.8 14.7 9.5 12.7 14.7 Median FCR (1.69) 11.2 14.4 16.4 10.9 14.4 16.4 10.7 14.3 16.4 High FCR (2.0) 13.6 17.5 19.8 13.3 17.4 19.8 13.0 17.3 19.8

Low FCR (1.63) Median FCR (2.25) High FCR (3.0) Low FCR (1.63) Median FCR (2.25) High FCR (3.0) 1.9 28.8 36.9 9.7 46.8 57.9 19.0 68.5 83.4 6.0 12.7 18.7 9.20 18.4 26.6 13.0 25.3 36.2

Low = 5th percentile, median = median, high = 95th percentile

Fig. 3 The proximate composition, energy content and phosphorus content (on as fed basis) of the different feed types, and categories. Bars represent mean SD. For any one parameter, different superscripts indicate signicant differences (P [ 0.05). The

commercial feeds are categorised, based on manufacturers recommendations on the stock size where for 4150 g, 20200 g and over 500 g sized sh are termed small feed (SF), medium feed (MF) and large feed (LF), respectively, and FMF refer to farm-made feeds

two types of feeds (SF and MF) are used mostly for 35 months of the culture period, and it is estimated that nearly 60% of the stock by weight is fed on rst two feed types, whilst larger sh are commonly fed with the third type (LF). The chemical composition of the different feed types vary to some extent, the most signicant being the lower protein content of LF, used for stock over 500 g (Fig. 3), with a protein content comparable to that of farm-made feeds. All commercial feeds had a signicantly lower level of lipid content compared to that in farm-made feeds,

which exceeded 10%, and similarly the amount of ash was lower in the former type of feeds. The energy content and phosphorus content did not differ signicantly between feed types (P [ 0.05). The data on feeds in the catsh farming sector collected by Phan et al. (2009) were not utilised in detail previously. However, the above study was linked to the present one and in the latter details pertaining to feed type efcacies were analysed in detail and presented. The FCR for commercial pellets and farm-made feed ranged from 1.0 to 3.0

123

Royal Swedish Academy of Sciences 2010 www.kva.se/en

AMBIO (2010) 39:504514

60
Farm made

509

50 40

Commercial

30 20 10 0 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3

FCR

Fig. 4 The distribution of FCR in the two feed types in the 97 surveyed farms (based on data from Phan et al. 2009)
1000
Commercial Farmmade

800

Yield (t/ha/crop)

600

400 200

Nutrient loadings from catsh farming for different food types and N and P content of diets were estimated for different FCRs (Table 1). Estimated nutrient loadings for farm-made feeds were substantially more variable than for commercial feeds, but median values were similar for N loadings. Nutrient loadings for commercial feeds ranged from 33.4 to 69.7 kg/t (median 46.0 kg/t) and 9.5 to 19.8 kg/t (median 14.4 kg/t), for N and P, respectively, while for farm-made feeds loadings ranged from 1.9 to 83.4 kg/t (median 46.8 kg/t) and 6.0 to 36.2 kg/t (18.4 kg/t) for N and P, respectively (Table 1). An increase in the proportion of contribution from Phase I increased the N loading due to the high protein content of the diet for smaller sh, but decreased in the P loading (Table 1). Decreasing the FCR reduced both the N and P loadings. The quantum of striped catsh production in the Mekong Delta grown on commercial feed and farm-made feed is not known. Since the vast majority of farms used commercially made feed, though a proportion of these also use farm-made feed (Phan et al. 2009) using the above median nutrient loadings for both commercial and farm-made feeds, in 2007 when 687,000 t of striped catsh was produced in the Mekong Delta, 31,60232,152 t N and 9,89318,274 t P were discharged, while in 2008, 50,364 51,240 t N and 15,76629,124 t P were discharged (Table 2).

Percent

DISCUSSION
1 1.5 2 FCR 2.5 3

Fig. 5 The scatter diagram of relationship of FCR to yield for the two feed types used (based on data from Phan et al. 2009)

(mean 1.69), and 1.3 to 3.0 (mean 2.25), respectively, which differed signicantly (P \ 0.001) from each other (Figs. 4, 5). The relationship between feed type and yield was insignicant (P [ 0.05), but the production cycle when using farm-made feed was usually 48 weeks longer. There was no discernible statistical relationship between yield and FCR for either type of feed (Fig. 5). Gross Estimates of Nitrogen and Phosphorus Discharge Often estimates of nutrient loading in aquaculture practices have been based on laboratory studies of single units of production, barring a few cases (Ackefors and Enell 1990, 1994). To date there has been no detailed study of the nutrient loading in discharge waters from striped catsh farms in the Mekong Delta, and for that matter for a culture practice of the scale of that of catsh, taken as a unit.

In this study, we have attempted to estimate nitrogen and phosphorus discharge from an aquatic farming system, which is one of the biggest globally, but concentrated in a relatively small geographic area of approximately 6,000 7,000 ha (Nguyen and Oanh 2009; Phan et al. 2009). To place it in perspective the current production from the sector, in this relatively small area, is equivalent to nearly 70% of total European aquaculture production. In our attempt to estimate N and P discharge from the striped catsh farming sector in the Mekong Delta we have taken into account the type of feed used, commercial or farm-made, even though the differences in the proximate composition of the two types of feed types are relatively minor, the FCRs, and aspects of feed usage (use of low protein content feed in the second phase of the grow-out cycle) in the different phases of the grow-out cycle. The main difference between the two feed types (commercial and farm-made) was the relatively high level of lipid in farm-made feeds. FCRs for catsh farming in the Mekong Delta range from 1.0 to 3.0 (mean 1.69), and 1.3 to 3.0 (mean 2.25) for commercial pellets and farm-made feed, respectively (Phan et al. 2009). These values, which appear to be generally

Royal Swedish Academy of Sciences 2010 www.kva.se/en

123

510 Table 2 N and P nutrients loadings for striped catsh production in the Mekong Delta in 2007 and 2008 Variable Production level (91,000 t): Commercial feeds Feed required (using median FCR of 1.69) (91,000 t) Nutrient loading (t) based on nutrient content of catsh carcass Farm-made feeds Feed required (using median FCR of 2.25) (91,000 t) Nutrient loading (t) based on nutrient content of catsh carcass 1,546 32,152 12,641 1,161 31,602 9,893 2007 687 N P

AMBIO (2010) 39:504514

2008 1,094.879 N 1,850 50,364 2,463 51,240 20,146 15,766 P

Nutrient loadings based on median values for sh fed either commercial or farm-made feeds calculated using a nutrient mass balance model (median values from Table 1) Table 3 N and P nutrient loadings for different farmed species Species Culture system Feed Discharge (kg/t) N Rainbow trout (Oncorhynchus mykiss) Silver perch (Bidyanus bidyanus) Channel catsh (Ictalurus punctatus) and Bluntsnout bream (Megalobrama amblycephala) Areolated grouper (Epinephelus areolatus) Bluen tuna (Thunnus maccoyii) Gilthead seabream (Sparus aurata) Common carp (Cyprinus carpio) Cage, raceway Pond, cage Pond/cage P Lanari et al. (1995), Ingram (1999), Bureau et al. (2003) Guo and Li (2003), Guo et al. (2009) Leung et al. (1999) Fernandesa et al. (2007) 17.8 Lupatsch and Kissil (1998) Watanabe et al. (1999), Jahan et al. (2002) Median values in present study Median values in present study Authority

Commercial 47.371.1 6.524.2 Commercial 130 120160

14.428.8 Gooley et al. (2000, 2001a, b) 2535

Cage Cage Cage Pond

Trash sh Fish Various

321 260502 30.986.0 8.526.4 14.4 26.6

Commercial 102.9

Striped catsh (Pangasianodon hypophthalmus) Pond Striped catsh (Pangasianodonhypophthalmus) Pond

Commercial 46.0 Farm-made 46.8

higher than FCRs for other, well established aquaculture nsh where FCRs near or less than 1.0 are often reported (e.g., Dumas et al. 2007). Commercial feeds for catsh analysed in this study may not have been fully optimised for catsh and, as suggested by Phan et al. (2009), their quality being highly variable. Indeed there have been no detailed studies into the nutritional requirements of catsh. Renement of feed formulations combined with improved feeding strategies (such as to reduce feed wastage) may improve FRC in catsh farming, and thereby reduce nutrient loadings in efuent water, as well enable the practices to be more economically viable (Department of Aquaculture 2008; Sub-Institute for Fisheries Economics and Planning in Southern Vietnam 2009). Also, a comparison of the amounts of N and P discharged in the production of a tonne of some selected cultured sh species (Table 3) indicates that at present striped catsh farming in the Mekong Delta compares favourably with other cultured species, irrespective of the type of feed used. This may be due to the relatively high

feed efciency of striped catsh, and by generally lower protein and higher carbohydrate content of catsh diets compared to other species (Yi et al. 2004). In comparison to published N and P content values for striped catsh (1.8% N and 0.21% P wet weight) (Yi et al. 2004), values for other species range from 2.2 to 3.4% and 0.39 to 1.20% of wet weight, for N and P, respectively (Lall 1991; Lupatsch and Kissil 1998; Gooley et al. 2001; Schreckenbach et al. 2001; Kiessling 2009). Nevertheless, nutrient loadings presented here for striped catsh farming are to be considered as preliminary estimates only and further, more detailed studies are warranted to rene and validate these values. From the view point of environmental well being it is also important to consider the total N and P discharges and the actual extent of overall nutrient loading from this rather large aquaculture farming system in the Mekong Delta. Based on values for commercial feeds, the total amount of N and P discharged from catsh farming in the Mekong Delta in 2007 and 2008 were estimated to be 31,602 t N

123

Royal Swedish Academy of Sciences 2010 www.kva.se/en

AMBIO (2010) 39:504514

511

and 9,893 t P, and 50,364 t N and 15,766 t P, respectively (Table 2). On the other hand, if the whole of the Mekong River striped catsh farming sector were to use farm-made feeds the total loading in 2007 and 2008 would be 32,152 and 51,240 t of N, and 12,641 and 20,146 t of P, respectively (Table 2). Thus it seems that the use of commercially compounded feeds is responsible for a very minimal, almost trivial, reduction of N, but greatly reduces the amount of P discharged. However, improved formulation guidelines for farm-made feeds may reduce this value. In an earlier pioneering study of two decades back, of Ackefors and Enell (1990), a comparable estimation was done for the entire Swedish salmonid industry (salmons and trouts). It was reported that the total yearly loading of N and P from the Swedish salmonid industry equated to 307 and 38 t, respectively, for a total production of 3,945 t of cultured sh (equating to 78 kg of N and 10 kg P, per tonne of produced sh). However, the authors also estimated the so called self purication derived by denitrication, biological uptake and sedimentation, reporting a nal estimate of a total load of 264 t of N and 35 t of P (hence, a reduction of roughly 14% of N and 8% of P). In a follow-up study (Ackefors and Enell 1994) by the same authors it was reported that these values were progressively decreasing with the optimisation of feed and feeding practices. Accordingly, and more recently (Kiessling 2009) it was clearly shown that, resulting from continuous improvement in feed regimes and feed composition, the average load of N and P (as kg/t of production) from Swedish sh farming has been signicantly reduced to values of approximating 55 and 7 kg/t for N and P, respectively. Comparing these data on the Swedish salmonid industry with those presented in this study for the stripped catsh it is interesting to note that, in general, N load is lower, whilst P load is higher for the current Vietnamese catsh industry. Furthermore, the latter being a relatively younger development there is a lot of scope for improvement and gains in nutrient discharge levels through the adoption of improved feeds and feed management practices, in conjunction with planned water intake and discharge into the environment especially through the adoption of Better Management Practices and a cluster approach. Admittedly, not all of the N and P are discharged directly into the Mekong River. Some catsh farms in the Mekong Delta (approximately 35%) discharge water directly onto rice elds or gardens (Phan et al. 2009), and some nutrients will be retained in the sludge of ponds or taken up by other biota in the environment. Phan et al. (2009) estimated that in 2007, when 683,000 t of catsh was produced in the whole of the Mekong Delta, 6.4 ML of water was used per t of production, which is about 10% higher than the estimates made previously (Bosma et al. 2009). On the above grounds it can be concluded that the total discharge of nutrients directly into the Mekong Delta

from catsh farming is substantially less than the total N and P loads estimated in this study, and the need for more accurate estimations taking into consideration the many variable factors in use and discharge of water from catsh farming practices will enable to obtain more rened estimates. The development of Better Management Practices (BMPs) and the application thereof through a cluster approach (NACA 2009), as had been successfully done in the case of shrimp farming in India (Umesh et al. 2009) will lead to further betterment in feed efcacies and water management regimes, such as through the introduction of water calendars in the catsh farming sector leading to signicant reductions in N and P discharge levels and thereby contribute to the sustainability of the sector and to improved environmental integrity. More detailed studies on the nutrient dynamics within the Mekong Delta, incorporating catsh farming and its interactions with other primary production sectors (especially rice farming), are required. In particular, in a similar fashion to the assessment of self purication of nutrient loads previously computed for salmonid (Ackefors and Enell 1990), studies on the fate of N and P from catsh farming, such as determining retention rates of N and P in the pond sediments and amounts captured by other farming practices (e.g., rice farming), will enable further renement of such estimates. These in turn will be useful for overall development planning of the sector as well as provide useful information to a holistic approach to assessing the river well being. Further more, available information suggests that this quantity of potential discharge from the striped catsh farming sector is relatively small compared to the potential run-off of N and P from fertilizers used in rice farming in the Mekong Delta. In fact, rice cultivation in the Mekong Delta used 170182 kg of fertilizer per sown acre of paddy, totalling 7.48 million kilogram in 2000 (Truong 2003). Further, it has been suggested that the water quality in the Mekong River between 2005 and 2008 has hardly changed compared to the period prior to the expansion of the catsh farming sector (Bosma et al. 2009). However, as pointed out by Phan et al. (2009), all this is not a matter for complacency, nor does it preclude the fact that further improvements in the manner that efuent is discharged are envisaged to be essential, such as through the adoption of BMPs (NACA 2009). Within this context, an appropriate and promising example is derived by the noteworthy progress of nutrient load reduction recorded for the salmonid industry (Kiessling 2009); an industry roughly 40 years older than the striped catsh farming in Vietnam. The striped catsh farming along the Mekong Delta and its tributaries occur upstream of 5060 km of the sea mouth. The Mekong River has the 10th highest ushing rate of all rivers in the world (van Zalinge et al. 2004). It is

Royal Swedish Academy of Sciences 2010 www.kva.se/en

123

512

AMBIO (2010) 39:504514 Corsin, F. 2007. Dialogue on catsh industry aims for sustainability through consensus and certication. Catch and Culture 13(2): 1819. dOrbcastel, E.R., J.P. Blancheton, T. Boujard, J. Aubin, Y. Moutounet, C. Przybyla, and A. Belaund. 2008. Comparison of two methods for evaluating waste of a ow through trout farm. Aquaculture 274: 7279. De Silva, S.S., and T.A. Anderson. 1996. Fish nutrition in aquaculture, 256. UK: Chapman and Hall. De Silva, S.S., and F.B. Davy. 2009. Aquaculture successes in Asia: Contribution to sustained development and poverty alleviation. In Success stories in Asian aquaculture, ed. S.S. De Silva, and F.B. Davy, 114. The Netherlands: Springer, IDRC, NACA. De Silva, S.S., and G.M. Turchini. 2008. Towards understanding the impacts of the pet food industry on world sh and seafood supplies. Journal of Agricultural and Environmental Ethics 21: 459467. Department of Aquaculture. 2008. Report on the implementation progress of the 2008 plan, and major solutions for implementation of the 2009 plan for the aquaculture sector. Ministry of Agriculture and Rural Development, 7 pp (in Vietnamese). Dudgeon, D. 2000. Riverine biodiversity in Asia: A challenge to conservation biology. Hydrobiologia 418: 113. Dumas, A., C.F.M. de Lange, and D.P. Bureau. 2007. Quantitative description of body composition and rates of nutrient deposition in rainbow trout (Onccorhynchus mykiss). Aquaculture 273: 139146. FAO. 2008. The state of world sheries and aquaculture 2008, 176. Rome: FAO. Fernandesa, M., P. Lauera, A. Cheshirea, and M. Angovec. 2007. Preliminary model of nitrogen loads from southern bluen tuna aquaculture. Marine Pollution Bulletin 54: 13211332. Globesh. 2009. Pangasius market report, 3 pp. Goddard, S. 1996. Feed management in intensive aquaculture. New York, USA: Chapman & Hall. 194 pp. Gooley, G.J., S.S. De Silva, P.W. Hon, L.J. McKinnon, and B.A. Ingram. 2000. Cage aquaculture in Australia: A developed country perspective with reference to integrated aquaculture development within inland Australia. In Cage aquaculture in Asia: Proceedings of the rst international symposium on cage aquaculture in Asia, ed. I.C. Liao, and C.K. Lin, 2137. Bangkok, Thailand: Asian Fisheries Society, Manila and World Aquaculture Society-Southeast Asian Chapter. Gooley, G.J., L.J. McKinnon, B.A. Ingram, and R. Gasior. 2001a. Multiple use of farmwater to produce sh, 98. Barton, ACT, Australia: Rural Industries Research and Development Corporation. Gooley, G.J., S.S. De Silva, B.A. Ingram, L.J. McKinnon, F.M. Gavine, and W. Dalton. 2001b. Cage culture of nsh in Australian lakes and reservoirsA pilot scale case study of biological, environmental and economic viability. In Reservoir and culture-based sheries; biology and management. Proceedings of the international workshop held in Bangkok, Thailand from 1518 February 2000. ACIAR Proceedings No. 98, ed. S.S. De Silva, 328346. Canberra, Australia: Australian Centre for International Agricultural Research. Guo, L., and Z. Li. 2003. Effects of nitrogen and phosphorus from sh cage-culture on the communities of a shallow lake in middle Yangtze River basin of China. Aquaculture 226: 201212. Guo, L., Z. Li, P. Xie, and L. Ni. 2009. Assessment effects of cage culture on nitrogen and phosphorus dynamics in relation to fallowing in a shallow lake in China. Aquaculture International 17: 229241. Ingram, B.A. 1999. A phosphorus model for trout farming in the Goulburn-Broken catchment. In Towards best practice in landbased salmonid farming: Options for treatment, re-use and

therefore, conceivable the nutrients discharged are carried to the South China sea rather rapidly, enriching the immediate waters. In the wake of climate change and mitigating measure of carbon sequestration, studies on the Amazon have demonstrated that nutrient discharge from rivers through increased diazotrophy could enhance carbon sequestration (Cooley et al. 2007; Subramanium et al. 2008). As such the striped catsh farming system in the Mekong Delta not only is of enormous socio-economic importance to the region, it is also currently responsible for relatively minor environmental impacts in term of N and P discharge, but also in the long run could help in sequester carbon in signicant quantities.
Acknowledgements This work was undertaken as a component of the Collaboration for Agriculture and Rural Development (CARD) program between the Governments of Vietnam and Australia, funded through the Australian Agency for International Development (AusAID). The current project is a component of the project Development of better management practices for catsh farming in the Mekong Delta (VIE 001/07). We are grateful to the nancial support provided by AusAID. Most of all, our thanks are due to numerous farmers who were very willingly forthcoming with information and provided access to their records unreservedly. We value their friendship and cooperation. We also wish our thanks to other colleagues who were involved in this project, in particular Drs. Nguyen Van Hao, Phan Than Lam and Thuy T.T. Nguyen.

REFERENCES
Ackefors, H., and M. Enell. 1990. Discharge of nutrients from Swedish sh farming to adjacent sea areas. AMBIO 19: 2835. Ackefors, H., and M. Enell. 1994. The release of nutrients and organic matter from aquaculture systems in Nordic countries. Journal of Applied Ichthyology 10: 225241. Bosma, R.H., C.T.T. Hanh, and J. Potting. 2009. Environmental impact assessment of the pangasius sector in the Mekong Delta. Wageningen University, 50 pp. Bureau, D.P., S.J. Gunther, and C.Y. Cho. 2003. Chemical composition and preliminary theoretical estimates of waste outputs of rainbow trout reared in commercial cage culture operations in Ontario. North American Journal of Aquaculture 65: 3338. Cho, C.Y., and D.P. Bureau. 1998. Development of bioenergetic models and the Fish-PrFEQ software to estimate production, feeding ration and waste output in aquaculture. Aquatic Living Resources 11: 199210. Cho, C.Y., J.D. Hynes, K.R. Wood, and H.K. Yoshida. 1991. Quantitation of sh culture wastes by biological (nutritional) and chemical (limnological) methods: The development of high nutrient dense (HND) diets. In Nutritional strategies and aquaculture waste. Proceedings of the rst international symposium on nutritional strategies in management of aquaculture waste, ed. C.B. Cowey, and C.Y. Cho. Guelph, Ontario: Fish Nutrition Research Laboratory, University of Guelph. Cho, C.Y., J.D. Hynes, K.R. Wood, and H.K. Yoshida. 1994. Development of high-nutrient-dense, low- pollution diets and prediction of aquaculture wastes using biological approaches. Aquaculture 124: 293305. Cooley, S.R., V.J. Coles, A. Subramanium, and P.L. Yager. 2007. Seasonal variations in the Amazon plume-related atmospheric carbon sink. Global Biogeochemical Cycles 21 (GB3014).

123

Royal Swedish Academy of Sciences 2010 www.kva.se/en

AMBIO (2010) 39:504514 disposal of efuent, ed. B.A. Ingram, 2641. Alexandra, VIC, Australia: Marine and Freshwater Resources Institute. Jahan, P., T. Watanabe, S. Satoh, and V. Kiron. 2002. A laboratorybased assessment of phosphorus and nitrogen loading from currently available commercial carp feeds. Fisheries Science 68: 579586. Kelly, L.A., J. Stellwagen, and A. Bergheim. 1996. Waste loadings from a freshwater Atlantic salmon farm in Scotland. Journal American Water Resource Association 32: 10171025. Kiessling, A. 2009. FeedThe key to sustainable sh farming. In: Fisheries, sustainability and development. Royal Swedish Academy of Agriculture and Forestry (KSLA). pp. 303322. Lall, S.P. 1991. Digestibility, metabolism and excretion of dietary phosphorus in sh. In Nutritional strategies and aquaculture waste, ed. C.B. Cowey, and C.Y. Cho, 2136. Guelph, ON, Canada: University of Guelph. Lanari, D., E. Dagaro, and R. Ballestrazzi. 1995. Dietary N and P levels, efuent water characteristics and performance in rainbow trout. Water Science and Technology 31: 157165. Leung, K.M.Y., J.C.W. Chu, and R.S.S. Wu. 1999. Nitrogen budget for the areolated grouper Epinephelus areolatus cultured under laboratory conditions and in open-sea cages. Marine Ecology Progress Series 186: 271281. Lupatsch, I., and G.W. Kissil. 1998. Predicting aquaculture waste from gilthead seabream (Sparus aurata) culture using a nutritional approach. Aquatic Living Resources 11: 265268. NACA. 2009. Better Management Practices (BMPs) for Striped Catsh (Tra catsh) Farming Practices in the Mekong Delta, Vietnam, 75 pp. http://library.enaca.org/inland/catshbmps/cat sh_bmp_v2.pdf.org/inland/catshbmps/. Naylor, R.L., R.J. Goldburg, H. Mooney, M. Beveridge, J. Clay, C. Folke, N. Kautsky, J. Lubchenco, J. Primavera, and M. Williams. 1998. Natures subsidies to shrimp and salmon farming. Science 282: 883884. Naylor, R.L., R.J. Goldburg, J. Primavera, N. Kautsky, M. Beveridge, J. Clay, C. Folke, J. Lubchenco, H. Mooney, and M. Troell. 2000. Effect of aquaculture on world sh supplies. Nature 405: 10171024. Naylor, R.L., S.L. Williams, and D.R. Strong. 2001. Aquaculturea gateway for exotic species. Science 294: 16551666. Naylor, R.L., R.W. Hardy, D.P. Bureau, A. Chiu, M. Elliot, et al. 2009. Feeding aquaculture in an era of nite resources. PNAS 106 (36): 1510315110. www.pnas.org/cgi/doi/10.1073/pnas.09 05235106. Neubacher, H. 2009. Pangasius: Creating a benchmark. Fishfarming International, June 2009, 17 pp. New, M. 1991. Compound feedsworld view. Fish Farmer, March/ April 1991, pp. 3946. New, M. 1997. Aquaculture and the capture sheries-balancing the scales. World Aquaculture 28: 1130. Nguyen, P.T., and D.T.H. Oanh. 2009. Striped catsh (Pangasianodon hypophthalmus) aquaculture in Viet Nam: An unprecedented development within a decade. In Success stories in Asian aquaculture, ed. S.S. De Silva, and F.B. Davy, 133150. The Netherlands: Springer, IDRC, NACA. Phan, L.T., T.M. Bui, T.T.T. Nguyen, G.J. Gooley, B.A. Ingram, H.V. Nguyen, P.T. Nguyen, and S.S. De Silva. 2009. Current status of farming practices of striped catsh, Pangasianodon hypophthalmus in the Mekong Delta, Vietnam. Aquaculture 296: 227236. Sapkota, A., A.R. Sapkota, M. Kucharski, J. Burke, S. MvKinenzie, P. Walker, and R. Lawrence. 2009. Aquaculture practices and potential human health risks: Current knowledge and future priorities. Environment International 34: 12151226. Schreckenbach, K., R. Knosche, and K. Ebert. 2001. Nutrient and energy content of freshwater shes. Journal of Applied Ichthyology 17: 142144.

513 Sodhi, N.S., L.P. Koh, B.W. Brook, and P.K.L. Ng. 2004. Southeast Asian biodiversity: An impending disaster. Trends in Ecology and Environment 19: 654660. Subasinghe, S., D. Soto, and J. Jia. 2009. Global aquaculture and its role in sustainable development. Reviews in Aquaculture 1: 29. Sub-Institute for Fisheries Economics and Planning in Southern Vietnam. 2009. Project on development planning for catsh production and consumption in the Mekong Delta up to 2010 and strategic planning up to 2020. Ho Chi Minh City, Vietnam: Department of Aquaculture, Ministry of Agriculture and Rural Development, 124 pp. Subramanium, A., P.L. Yager, E.J. Carpenter, C. Mahaffey, K. rkman, et al. 2008. Amazon River enhances diazotrophy and Bjo carbon sequestration in the tropical North Atlantic Ocean. Proceedings of the National Academy of Sciences of the United States of America 105: 1046010465. Truong, V.T. 2003. A draft inception report on integrated assessment of trade liberalization in rice sector, Vietnam. HaNoi, Vietnam: UNEP. 28 pp. Umesh, R.N., A.B. Chandra Mohan, G. Ravibabu, P.A. Padiyar, M.J. Phillips, C.V. Mohan, and B. Vishnu Bhat. 2009. Implementation of better management practices by empowering small-scale farmers through a cluster-based approach: The case of shrimp farmers in India. In Success stories in Asian aquaculture, ed. S.S. De Silva, and F.B. Davy, 4365. The Netherlands: Springer, IDRC, NACA. van Zalinge, N.P., P. Degen, P. Pongsiri, S. Nuov, J.G. Jensen, V.H. Nguyen, and X. Choulamany. 2004. The Mekong River system. In Proceedings of the second international symposium on the management of large rivers for sheries, ed. R.L. Welcomme, and T. Petr, 335357. Bangkok, Thailand: RAP Publication 2004/16, Vol. 1. FAO Regional Ofce for Asia and the Pacic. Watanabe, T., P. Jahan, S. Satoh, and V. Kiron. 1999. Total phosphorus loading onto the water environment from common carp fed commercial diets. Fisheries Science 65: 712716. Wijkstrom, U.N., and M.B. New. 1989. Fish for feed: A help or a hindrance to aquaculture in 2000? INFOFISH International 6(89): 4852. World Wild Life Mediterranean Programme. 2005. Risk on local sh populations and ecosystems posed by the use of imported feed sh by the tuna farming industry in the Mediterranean. Barcelona, Spain, 11 pp. Yi, Y., D.R. Yuan, N.T. Phuong, T.Q. Phu, C.K. Lin, and J.S. Diana. 2004. Environmental impacts of cage culture for catsh in Hongngu, Vietnam. In Twenty-rst annual technical report, ed. R. Harris, and EgnaH CourterI, 157168. Oregon State University, Oregon, USA: Aquaculture CRSP.

AUTHOR BIOGRAPHIES
Sena S. De Silva (&) is the Director General of Network of Aquaculture Centers in Asia-Pacic and also an Honorary Professor, School of Life and Environmental Sciences, Deakin University, Victoria, Australia. Address: Network of Aquaculture Centres in Asia-Pacic (NACA), PO Box 1040, Kasetsart Post Ofce, Bangkok 10903, Thailand. Address: School of Life and Environmental Sciences, Deakin University, Warrnambool, VIC, Australia. e-mail: sena.desilva@enaca.org Brett A. Ingram is a senior scientist of the Fisheries Victoria, Department of Primary Industries, Victoria, Australia. Address: Department of Primary Industries, Fisheries Victoria, Melbourne, VIC, Australia. e-mail: brett.ingram@dpi.vic.gov.au

Royal Swedish Academy of Sciences 2010 www.kva.se/en

123

514 Phuong T. Nguyen is Professor of Aquaculture of the Faculty of Fisheries and Aquaculture, University of CanTho, Vietnam. Address: College of Aquaculture and Fisheries, Can Tho University, Can Tho, Vietnam. e-mail: ntphuong@ctu.edu.vn Tam M. Bui is a Senior Lecturer in Aquaculture, of the Faculty of Fisheries and Aquaculture, University of CanTho, Vietnam. Address: College of Aquaculture and Fisheries, Can Tho University, Can Tho, Vietnam. e-mail: bttam@ctu.edu.vn

AMBIO (2010) 39:504514 Geoff J. Gooley is a Senior Scientist of the Fisheries Victoria, Department of Primary Industries, Victoria, Australia. Address: Department of Primary Industries, Fisheries Victoria, Melbourne, VIC, Australia. e-mail: Geoff.gooley@dpi.vic.gov.au Giovanni M. Turchini is a Lecturer of the School of Life and Environmental Sciences, Deakin University. Address: School of Life and Environmental Sciences, Deakin University, Warrnambool, VIC, Australia. e-mail: giovanni.turchini@deakin.edu.au

123

Royal Swedish Academy of Sciences 2010 www.kva.se/en